This project is a continuation of studies performed in the previous Projects 1 and 2. We obtained evidence that helper T cell responses and antibody responses to the MUC1 tumor antigen peptide are reduced in MUCITg mice compared to WT mice but when the peptide carried 0-linked GalNac residues, representing hypoglycosylated tumor form of MUC1, MUCITg mice responded equally well as the WT mice. We hypothesize that the difference in responses to the MUC1 peptide versus glycopeptide presented by DC is due to tolerance because MUC1 peptide is perceived as "self," while tumor-specific glycopeptide represents abnormal self and thus is perceived as foreign. Similar state of tolerance may characterize MUC1 peptide specific T cells in patients and thus a related hypothesis is that MUC1 glycopeptide is a better vaccine candidate than the peptide that has been used to date. Our third hypothesis is that immune parameters (biomarkers) of efficacy exist as complex signatures of immune activation of DC and T cells and our goal is to identify those that can predict anti-tumor efficacy of MUC1 vaccines. We propose to test these hypotheses in human MUC1 transgenic mice and MUC1 peptide and glycopeptide specific TCR transgenic mice, as well as in rhesus macaques using rhesus MUC1 sequences. We propose the following specific aims: Specific Aim 1: We will determine what controls differences in immune responses to the self/tumor antigen MUC1 in MUCITg mice when MUC1 peptide (self?) versus MUC1 glycopeptide (foreign?) are used as antigens and targeted to adjuvant-activated DC. Specific Aim 2: We will test different MUC1 vaccines in rhesus macaques The proposed experiments will draw on observations already made or to tte made in the Projects 1 and 3 that identify important parameters of T cell and DC activation. The ultimate goal of Project 2 is to define the next generation of MUC1 vaccines.capable of eliciting an immune response with a predetermined signature of anti-tumor efficacy.